The neostriatum is the target of most afferent inputs to the basal ganglia, including the enormous projection from the cerebral cortex and the important modulatory dopamine input from the substantia nigra. These input pathways converge on the principal cells of the neostriatum, the spiny cells, which project to the basal ganglia output structures. Thus these afferents and the mutual interconnections of the spiny cells make up the primary input output circuit of the striatum and a major portion of the circuitry of the basal ganglia. Being such a prominent anatomical feature, the inhibitory interconnections between the striatal spiny neurons have been assumed to be a critical functional component, and have become a major element in most theories of basal ganglia function. These theories include those used to explain the actions of l-dopa, transplantation therapy and pallidotomy in therapy for Parkinsonism. Recent direct tests in adult animals have shown this interconnection between spiny cells to be weak to the point of being experimentally undetectable. This application will test possible explanations for this mismatch between anatomical (and theoretical) expectations and physiological reality. Special attention will be paid to the possibility that mutual inhibition among spiny neurons may only occur during plastic changes in the connections of the corticostriatal system, and that it is modulated off (e.g. by an anti-Hebbian synaptic plasticity mechanism) when the correlations among striatal spiny neurons are low (as they are when tested under most experimental circumstances). This explanation has been offered as a possible mechanism based on theoretical work arising from neural network theory. It predicts that mutual inhibitory connections among spiny neurons will be experimentally demonstrable during development. This prediction will be tested using dual whole cell recording in slices of the rat neostriatum during development. In addition to these tests, anatomical measurements will be used to estimate the connectivity of the mutual inhibitory pathway for comparison with physiological measurements. This latter will allow us to determine whether there is an actual mismatch between physiological and anatomical estimates of connectivity (i.e. non-functional synaptic connections).